I am studying the wellposedness of the Burgers equation with initial data:
\begin{align}\label{eq:BP} \begin{cases} \partial_t u + u \partial_x u = 0, & t >0, \quad x \in\mathbb{R}, \\ u(0,x) = u_0(x), & x \in \mathbb{R}, \end{cases} \end{align}
The definition of the classic entropy condition for the initial value problem of this equation says that this due satisfies the following two assumptions:
- (Integral equation) $$ \int_{0}^{\infty} \int_{0}^{\infty}\left(u \varphi_{t}+\frac{u^{2}}{2} \varphi_{x}\right) \mathrm{d} x \mathrm{~d} t+\left.\int_{0}^{\infty} u_{0} \varphi \mathrm{d} x\right|_{t=0}=0 $$ for all test function.
- (Entropy condition) $$ u(x+z, t)-u(x, t) \leqslant C\left(1+\frac{1}{t}\right) z $$ for some $C\geq 0$ and almost all $x,\,z\in \mathbb{R}^+,\, t>0$.
However, if now I consider \begin{align} \begin{cases} \partial_t u + u \partial_x u = G(u), & t >0, \quad x \in\mathbb{R}, \\ u(0,x) = u_0(x), & x \in \mathbb{R}, \end{cases} \end{align} How can I define the entropy condition in this case ? Is it the same that the case before? Why?
A good starting point may be chapter 5 of Dafermos' book, where I quote from the abstract:
This deals with classical solutions first. Later weak solutions are also discussed. You can read on companion balance laws in Chapter 1.4, 1.5 of the aforementioned book.